Considerable evidence suggests that obsessive compulsive disorder (OCD) involves abnormal function in frontal brain circuits, specifically the cortico-striatal-pallidal-thalamic (CSPT) loops. Neuroimaging studies demonstrate excessive activity in some of the limbic-related components of the loops (orbitofrontal cortex [OFC], anterior cingulate cortex [OFC], and caudate nucleus [CN]), but the significance of these findings remains unclear. Aberrant functional activity may help localize dysregulated circuitry, but it could also reflect responses to abnormalities elsewhere. However, brain mapping studies have demonstrated that medial frontal structures, such as the ACC, play a central role in detecting errors. This psychological process appears to be disturbed in patients with OCD, who tend to over-focus on not making mistakes and avoiding potential bad outcomes. Preliminary data shows hyperactivity of an electrophysiological potential thought to be generated by the ACC, the error-related negativity, in patients with OCD. A critical function of the limbic CSPT loops (especially the OFC and CN) also involves the evaluation and anticipation of gains and losses, processes which determine the significance of possible actions and guide choice. This proposal will work from the theory that in patients with OCD, common thoughts acquire a pathological significance because the brain (through limbic CSPT circuits) over-estimates potential or actual losses. Functional magnetic resonance imaging and event-related potential studies will probe these CSPT circuits with simple tasks known to elicit neural responses in these brain areas, without eliciting symptoms of OCD. Two experiments will test hypotheses that OCD consists of abnormal brain activity in these regions during the commission of errors and when they anticipate and respond to potential monetary losses, allowing us to evaluate a cluster of hypotheses about the functional anatomy of OCD. Patients with OCD will be studied in medicated, relatively unsymptomatic samples, as well as in unmedicated, relatively symptomatic samples in order to identify trait-related components of the illness. The patient samples will be compared with psychiatricallyhealthy subjects and medicated, previously depressed patients. Successful completion of the project should lead to an improved understanding of the role of specific brain regions in generating the symptoms of OCD.